IBM Support

IC93289: WEBSPHERE MQ R530 ON HP-NSS CLUSTER PUB SUB MESSAGES FAIL WITH AMQ9538 - COMMIT CONTROL ERROR

Subscribe

You can track all active APARs for this component.

 

APAR status

  • Closed as program error.

Error description

  • When using a WebSphere MQ r531 queue manager running on HP-NSS
    which is in a cluster with pub-sub on other platforms the
    following type of message is received:
    Date  time  Process(x,process id y $number) User(user) Program (
    amqrmppa_r )
    AMQ9538: Commit control error.
    EXPLANATION:
    An error occurred when attempting to start commitment control.
    Either exception '' was received when querying commitment
    status, or commitment control could not be started.
    An FDC similar to this may also be produced:
    | WebSphere MQ First Failure Symptom Report
    | Date/Time         :-
    | Host Name         :- w
    | PIDS              :- 5724A3901
    | LVLS              :- 531.x
    | Product Long Name :- WebSphere MQ for HP NonStop Server
    :
    | Probe Id          :- CS075003
    | Application Name  :- MQM
    | Component         :- craWriteSync
    :
    | UserID            :- xxx,yyy (MQM.MANAGER)
    | Program Name      :- /usr/ibm/wmq/opt/mqm/bin/amqrmppa_r
    :
    | QueueManager      :-  QMGRNAME
    | Major Errorcode   :-  rrcE_COMMIT_CONTROL_ERROR
    :
    | Probe Type        :  MSGAMQ9538
    | Probe Description :- AMQ9538: Commit control error.
    :
    | Comment1          :- Error 2232 returned from
    lpiSPIHPNSSTxInfo
    

Local fix

Problem summary

  • If a batch containing persistent messages does not result in a
    physical MQPUT operation, the MCA does not start a TMF
    transaction.
    At the point the batch commits, the commit logic assumes there
    will always be a transaction in progress. Since this is not the
    case if no physical put operation has occurred, the commit
    attempt fails.
    In the case where a clustered subsub message arrives, the MCA
    attempts to DLQ the message, but since it is maked as
    MQRO_DISCARD,  the DLQ operation is suppressed.
    

Problem conclusion

  • If an incoming persistent message is discarded by reason of
    MQRO_DISCARD being set, and there is no TMF transaction in
    progress the MCA needs to start one.
    

Temporary fix

Comments

APAR Information

  • APAR number

    IC93289

  • Reported component name

    WEBS MQ NSS ITA

  • Reported component ID

    5724A3902

  • Reported release

    530

  • Status

    CLOSED PER

  • PE

    NoPE

  • HIPER

    NoHIPER

  • Special Attention

    NoSpecatt / Xsystem

  • Submitted date

    2013-06-21

  • Closed date

    2013-06-28

  • Last modified date

    2013-06-28

  • APAR is sysrouted FROM one or more of the following:

  • APAR is sysrouted TO one or more of the following:

Fix information

  • Fixed component name

    WEBS MQ NSS ITA

  • Fixed component ID

    5724A3902

Applicable component levels

[{"Line of Business":{"code":"LOB36","label":"IBM Automation"},"Business Unit":{"code":"BU053","label":"Cloud & Data Platform"},"Product":{"code":"SSFKSJ","label":"WebSphere MQ"},"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"5.3"}]

Document Information

Modified date:
17 November 2020